CompuServe Thread

#Flame IPAS

4 messages in this thread
#103201From: MARK H. HOUSERMay 21, 1994 2:55 PM
Sorry, but I think I'm still blowing it. If you load the sample project F_JETS.3DS, make a dummy object, link everything to it, goto the last frame and move the dummy about 30 or so units, the flame shoots way out the back. It seems as though the faster the object moves, the farther the flame rips out the back. In a seperate project I tried, a jet moving along at 500+ mph had 300 feet of flame coming out the back, while at about 25 mph it emitted a nice propulsion look. Help!??? – Robbo
#103210From: M. G. BATCHELORMay 21, 1994 4:51 PM
Mark, PMJI again, but after you moved the dummy with everything linked to it, did you go back into the AXP dialog and click OK to reset the secondary motion data ? If not, you should have received an error message upon rendering. Also, what is the length of the animation, and what is the particle life span ? I'll try to duplicate it on my system, also. BTW, I'm not affliliated with the YG. BILL
#103216From: MARK H. HOUSERMay 21, 1994 5:41 PM
Yes, I re-edited the settings, and clicked on Ok. I loaded the F_JETS.3DS sample, modified none of their existing settings, added the dummy and some motion, and rendered.
#103255From: Yost GroupMay 21, 1994 8:33 PM
You're not going insane. Here's what's going on. The Secondary Motion Data for the flame is being correctly generated to simulate what would happen if the flame source actually was transformed that much over that timeframe. It actually would be correctly if you had something like a torch that moved that fast (without getting blown out). However, jet engine dynamics work much differently than a torch… and therein lies the rub. We made the SMD dynamics in flame work as if you're dealing with a "natural" flame, like a torch. A jet engine creates a constant-velocity directed flame source, completely unaffected by the physics of how fast or in what direction the jet is going. The bottom line is that we should've put a switch into the Flame (and Vapor) settings dialogs that turn OFF Secondary Motion Data. (sound of hand hitting head) I now realize that we've got to do that, and we'll do a rev sometime in June that registered owners will be able to get off the Schreiber BBS that incorporates that switch. There are a few workarounds for you now, however. 1) The shorter the flame lifespan, the less effect that velocity will have on flame length (since SMD is calculated as the difference between the emitters over the lifespan period). So, a lifespan of 1 or 2 frames will give you the best shot at mitigating the SMD effect (plus, with a jet engine, that kind of frenetic motion will work fine). In addition to that, you'll have to keep the jet moving slow enough to keep the flames from getting out of hand… 1000 units over 200 frames will be way too much, for example. You'll have to scale your scene both in time and space to keep the SMD from getting carried away. 2) If you don't mind sitting next to the keyboard and hitting the Enter key after every frame, and if you need to have the temporal/spatial scaling of your scene very accurate (until we come up with the next rev of flame.axp, anyway), you can set all of your flame parameters BEFORE doing any of the jet keyframing, with any lifespan you want. Then, don't go back into the settings dialog AFTER you keyframe the sequence, to keep your SMD out of synch with what it really ought to be. The motion of the flame won't change over time, but you WILL have to hit the enter key on each frame to get past the error saying that the SMD is out of synch. Sorry about my oversight re jet engines… I think you'll be happy with the way we handle it in the 1.1 rev in June though. It'll be a simple button that just turns off SMD. (sound of hand hitting head AGAIN) Thanks for bringing it up, – G